226 M. L. FULLER THE SEA MILLS OF CEPHALONIA 



dilution, a similar working head might result from a depth of less than 

 50 feet, while with 50 per cent dilution only 25 feet would be necessary. 



For the production of circulation, by differences in density due to 

 dilution, no particular arrangement of the fissures is necessary and no 

 great depth is required, it being possible with a dilution of 25 per cent 

 to develop a head sufficient to produce circulation at a depth of less than 

 50 feet, or entirely above the zone controlled by interior temperatures. 



InevUdbleness of dilution. — The studies of the waters of deep mines 

 and tunnels on the one hand and of the underground waters in sediment- 

 ary rocks as' developed by deep borings on the other tend to show that, 

 except in the case of dense and fracture-free rocks, all formations are 

 commonly saturated from the water table down to considerable depths, 

 the water movement, according to the laws of underground waters, 

 which — thanks to King, Slicbter, and others — are almost as well known 

 as those of surface streams, being toward the lowest point of possible 

 leakage, which in the case of Cephalonia must be toward the sea. Cav- 

 ernous limestones are no exceptions to the rule of saturation, nor to 

 the normal seaward movement of the water, and there is little reason to 

 doubt that the rock passages, with which the Cephalonia limestone 

 abounds, are filled in a large measure with fresh waters moving toward 

 the sea. The prevalence of this movement near the sea mills is rendered 

 more probable by their position near the apex of a large reentrant bay, 

 the gulf of Argostoli, toward which the subterranean as well as the sur- 

 face drainage would naturally tend to converge. Besides this under- 

 ground drainage from the land, considerable amounts of fresh water 

 must reach the subterranean passages through sinks, joints, or crevices, 

 which are a characteristic feature of the limestone in the immediate 

 vicinity, the rugged fissured surface of which has already been noted. 

 ISTot only is there in all probability a considerable circulation through 

 the limestone, but there is reason to believe that the network of passages 

 is so complete that the dilution of any salt water entering the rock is in- 

 evitable. Besides the bedding and joint planes, which doubtless deter- 

 mined the location and controlled the formation of many of the passages, 

 there are many indications that earthquakes have done much to unite 

 them into a connected system of openings. Elsewhere in Greece earth- 

 ouake shocks have produced far-reaching effects upon the underground 

 drainage, old spring-? being destroyed in some cases and doubled in vol- 

 ume in others : new springs have broken out, and the waters of old springs 

 have changed in character and temperature. The Cephalonia region is 

 likewise one of violent earthquakes, some of disastrous severity having 



